Test Manager for Anomalous Log Subsequence Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manual selection of test cases for information handling systems is inefficient and unreliable due to the large number of potential test cases and types of systems, leading to potential missed anomalous results during testing operations.

Innovation Solution

A test manager uses an ensemble of prediction models to detect anomalous subsequences in log information, generating predictions and determining the next best test case by comparing processed subsequences with other test cases, thereby expanding testing coverage and improving efficiency and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual selection of test cases is used, then testing operations can be performed, but efficiency and reliability are insufficient due to the large number of potential test cases

Engineering Contradiction:

Improvetesting efficiencyVSAvoidtesting reliability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs self-service by automatically analyzing log information and determining the next best test cases without manual intervention. The test manager autonomously processes test results, identifies anomalies, and selects subsequent test cases based on learned patterns, eliminating the need for manual test case selection while improving both efficiency and reliability.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical manual process of test case selection with an automated machine learning system. The test manager uses trained models to analyze log information and automatically determine test cases, substituting human manual operations with an intelligent automated system that improves consistency and reliability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If the number of test cases is increased to improve coverage, then testing completeness improves, but the complexity of managing and selecting test cases increases

Engineering Contradiction:

Improvetesting coverageVSAvoidtest case management complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The test manager acts as an intermediary between the large set of available test cases and the testing system. It uses trained machine learning models to process log information and intelligently select the most appropriate next test cases, managing the complexity of large test case sets without requiring manual intervention.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system changes the parameter of test case selection from manual human decision-making to automated machine learning-based selection. By transforming the selection criterion into an automated process that analyzes log information patterns, the system can handle large numbers of test cases efficiently while maintaining high coverage.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If automated test case selection is implemented, then testing efficiency improves, but the complexity of the testing system increases

Engineering Contradiction:

Improvetesting efficiencyVSAvoidtesting system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary action by pre-training machine learning models on historical test data before actual testing operations. The test manager is trained using collected log information and test results in advance, so that during actual testing, it can efficiently and automatically select test cases without requiring complex real-time decision-making infrastructure.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11513927B1Method and system for performing testing operations for information handling systems

Publication Date: 2022.11.29 DELL PROD LP

AI summary

Techniques described herein relate to a method for performing testing operations for information handling systems. The method includes obtaining a test case from an information handling system; in response to obtaining the test case: obtaining log information associated with the test case from the information handling system; performing data preparation to generate processed subsequences using the log information; applying a plurality of prediction models to the processed subsequences and training data to generate anomalous subsequence predictions; generating ensemble anomaly scores and severity indexes associated with the processed subsequences using the anomalous subsequence predictions and the processed subsequences; making a determination that the ensemble anomaly scores and severity indexes associated with the processed subsequences result in detection of an anomalous subsequence; and in response to the determination: determining a next best test case associated with the anomalous subsequence; and initiating performance of the next best test case.